Abstract
Purpose
The majority (73%) of orthopaedic surgeons in Canada prefer using semitendinosus-gracilis (STG) autograft for ACL reconstruction. However, there is large variation in tendon size between individuals which makes pre-operative estimation of graft size unpredictable. Inadequate graft size may require an alternative source of graft tissue that should be planned prior to surgery. The purpose of this study is to determine if clinical anthropometric data and MRI measurements of STG tendons can be used to predict hamstring graft size.
Method
One-hundred and fourteen patients with ACL deficiency awaiting reconstruction using hamstring autograft were retrospectively evaluated. The following information was obtained from patient charts: height, weight, body mass index (BMI), age, and gender. Cross-sectional area (CSA) of gracilis (G) and semitendinosus (ST) tendons were determined from pre-operative MRI scans using NIH ImageJ analysis software. Actual STG graft diameters were obtained from operative reports. Correlations between patient height, weight, BMI, age, gender, ST-CSA, G-CSA, STG-CSA and intraoperative graft size were calculated to determine the association between these variables. Multiple stepwise regression was performed to assess the predictive value of these variables to intraoperative graft diameter. In addition, three investigators with no radiological experience made independent measurements of the ST and G tendons to determine the inter-rater reliability (ICC) of MRI measurements.
Results
All variables were independently correlated with intraoperative graft size (p<0.001). However, based on multiple stepwise regression analysis, only models including STG-CSA (r2=.212; p<.001); STG-CSA and sex (r2=.285; p<.001); and STG-CSA, sex and weight (r2=.294; p<.001) were found to be significant predictors of graft size (when co-variation in other factors was controlled). Inter-class correlation coefficients demonstrated very high agreement between raters for measurements of the ST, G and STG (.816, .827, .863, respectively).
Conclusion
Measurement of tendon CSA from MRI images is very reliable. A model including STG-CSA, sex and weight was found to be strongly predictive of hamstring graft diameter for ACL reconstruction. This model may enhance our ability to predict adequate graft size and identify instances that other graft tissues may be a better option. The results of this study may improve pre-operative planning for ACL reconstruction.
